J Oral Maxillofac Surg 68:144-148, 2010

The Efcacy of a Topical Anesthetic Gel in the Relief of Pain Associated With Localized Alveolar Osteitis
Corey C. Burgoyne, DMD,* James A. Giglio, DDS, MEd, Sarah E. Reese, BS, Adam P. Sima, MA, and Daniel M. Laskin, DDS, MS
Purpose: This prospective randomized clinical study assessed the efcacy of pain control for postex-

traction alveolar osteitis comparing the use of eugenol on a gauze strip versus a thermosetting gel containing 2.5% prilocaine and 2.5% lidocaine. Patients and Methods: Thirty-ve patients who presented with postextraction alveolar osteitis were randomly assigned to either a control group or test group. After irrigation of the extraction site with normal saline solution, the control patients were treated with eugenol on a gauze strip placed in the socket and the test patients were treated with the thermosetting gel placed directly into the socket. All patients were given a series of visual analog scales to record their pretreatment pain and post-treatment pain at 5, 10, and 15 minutes and then at 1-hour increments during waking hours for the next 48 hours. They were also given a prescription for an analgesic to use for breakthrough pain during the 48-hour period, if necessary, and instructed to record the dose and timing of any pain medication taken. All patients were seen for follow-up at 48 hours after treatment. Results: The mean pretreatment pain score was 6.72 on a scale ranging from 1 to 10 for the eugenol group and 6.37 for the prilocaine-lidocaine group (SE, 0.46), and the 2 groups were not different (P .62). In the immediate post-treatment period (0-15 minutes) the pain levels were signicantly reduced in both groups (Ps .001). However, the thermosetting gel produced a signicantly greater reduction in pain (mean, 3.23; SE, 0.62) than the eugenol (mean, 4.83; SE, 0.43) (P .022). Over the next 48 hours, the pain level was nominally less with the thermosetting gel, but this difference was not statistically signicant (Ps .2). Conclusion: Although the efcacy of the 2 treatments was not signicantly different, the nominal superiority and ease of using the thermosetting gel warrant further investigation. 2010 American Association of Oral and Maxillofacial Surgeons J Oral Maxillofac Surg 68:144-148, 2010 Over 100 million teeth are extracted yearly in the United States.1 One of the most common postoperative complications after the extraction of permanent teeth is localized alveolar osteitis (AO), also commonly known as dry socket.2 The incidence of AO is approximately 1% to 3% among all dental extractions, corresponding to approximately 100,000 to 300,000 patients per year in whom this painful postoperative complication will develop.
Received from Virginia Commonwealth University, Richmond, VA. *Formerly, Chief Resident, Department of Oral and Maxillofacial Surgery, School of Dentistry; and Currently, Private Practice, Richmond, VA. Professor, Department of Oral and Maxillofacial Surgery, School of Dentistry. Research Assistant, Department of Biostatistics, School of Medicine. Research Assistant, Department of Biostatistics, School of Medicine.

There are a variety of denitions used in the literature for the clinical diagnosis of AO. Blum,2 in a critical review of AO, attempted to establish the standardized denition as postoperative pain in and around an extraction site, which increases in severity at any time between 1 and 3 days after the extraction, accompanied by a partially or totally disintegrated blood clot within the alveolar socket with or without halitosis.
Professor and Chairman Emeritus, Department of Oral and Maxillofacial Surgery, School of Dentistry. Address correspondence and reprint requests to Dr Laskin: Department of Oral and Maxillofacial Surgery, School of Dentistry, Virginia Commonwealth University, PO Box 980566, Richmond, VA 23298-0566; e-mail: dmlaskin@vcu.edu
2010 American Association of Oral and Maxillofacial Surgeons

0278-2391/10/6801-0024$36.00/0 doi:10.1016/j.joms.2009.06.033

144

BURGOYNE ET AL

145 and absence of a clot in the socket on clinical examination. Exclusion criteria for the study included being immunosuppressed or currently taking immunosuppressant drugs; using steroids; having used nonsteroidal anti-inammatory drugs within 4 hours before being examined; taking drugs associated with drug-induced methemoglobinemia; having type 1 or type 2 diabetes; having a glucose 6-phosphate dehydrogenase deciency; having an allergy to eugenol, lidocaine, prilocaine, acetaminophen, or codeine; being pregnant; and being unable to provide informed consent. Those patients who met the diagnostic and other criteria were placed into a control group or an experimental group by use of a randomization table. There were 5 males and 15 females (mean age, 33 years; age range, 19-55 years) in the control group and 8 males and 7 females (mean age, 27 years; age range, 17-58 years) in the treatment group. The mandibular third molar was the most common tooth involved (21 of 35 [63%]), but there were also cases involving the mandibular left rst molar (n 5), the mandibular right rst molar (n 3), the mandibular right second molar (n 3), and the maxillary right and left second premolars and the maxillary left second molar (n 1 each). All patients were blind to which topical medication was used. Patients were initially asked to complete a visual analog scale (VAS) ranging from 0 (no pain at all) to 10 (the most pain imaginable) to measure their pain. The extraction site was then irrigated with room-temperature normal saline solution. The control group was treated with eugenol on a plain gauze strip placed in the extraction site. In the treatment group a gel containing 2.5% prilocaine, 2.5% lidocaine, poloxamer 188 and 407 (thermosetting agents), hydrochloric acid, and puried water (Oraqix; Dentsply Pharmaceutical, York, PA) was dripped into the extraction site up to the crest of the alveolus with a syringe. The thermosetting gel is liquid at room temperature and congeals at body temperature. On average, the volume of gel placed into the extraction site was less than 1 mL. All the patients were given a diary containing a series of VASs and asked to initially assess their pain at 5, 10, and 15 minutes after treatment. They were also given detailed instructions on how to complete the pain diary by recording the level of pain hourly on the VASs during waking hours for the next 48 hours. All patients were provided with a written prescription for 24 tablets of acetaminophen and codeine, 30 mg, taking 1 or 2 tablets every 4 hours if needed for pain. They were told to record in the diary the number of tablets used and the time when they were taken. All patients were told to return to the clinic after 48 hours for examination and for removal of any packing that had been placed. Those patients who still had pain were re-treated with the original medication.

The etiology of AO has not been established. However, the proposed theories include increased brinolysis with premature breakdown of the blood clot within the extraction site, surgical trauma, bacterial contamination, nutritional deciencies, mechanical dislodgement of the clot, compromised vascularity, tobacco use, and use of oral contraceptives.2 On the basis of these concepts, it is likely that the etiology is multifactorial. There are a multitude of studies that have examined ways to prevent AO. These have included antibacterial and antiseptic agents, antibrinolytics, steroids, and clot-supporting agents.3-5 However, despite the various measures that have been taken to prevent AO, it is still a common occurrence. In such cases the treatment is directed toward relief of pain until normal healing can take place. Various commercially available as well as individually prepared dressings have been used for this purpose. These have included antibacterial medications, obtundents, and topical anesthetics, as well as various combinations of these agents. Controversy exists about the placement of a physical dressing rather than a paste into the alveolus, and no studies have been performed that specically investigated the incidence of potential side effects and tissue damage arising from the placement of an intra-alveolar dressing. However, case reports describing local complications have been reported in the literature, and it is generally acknowledged that dressings delay the healing of the extraction socket.6-9 Because AO is essentially a pain problem, it seems logical that a topical anesthetic might be used for this purpose. In 1995 Betts et al10 studied the placement of viscous 2% lidocaine jelly versus a placebo into the socket and showed a short-term reduction in pain. Another study in 1995, by Donaldson and Meechan,11 assessed the topical efcacy of 5% lidocaine versus a 5% lidocaine-prilocaine mixture and found greater patient comfort with the latter combination. However, the problem with the use of a topical anesthetic is the need to retain the medication in the socket without the use of a carrier such as a gauze strip or resorbable material. This study investigated the use of a topical anesthetic combination that forms a gel when placed in the socket and therefore does not require a carrier.

Patients and Methods

Patients who presented to the oral surgery clinic at the Department of Oral and Maxillofacial Surgery, School of Dentistry, Virginia Commonwealth University (Richmond, VA), with complaints of postextraction pain consistent with AO were solicited to participate in this voluntary research study, which was approved by the institutional review board. The criteria used to establish the diagnosis were pain of increasing severity starting 2 to 3 days after extraction

146

TOPICAL ANESTHETIC AND ALVEOLAR OSTEITIS

FIGURE 1. VAS pain scores over time in the 2 groups with observed data. Burgoyne et al. Topical Anesthetic and Alveolar Osteitis. J Oral Maxillofac Surg 2010.

The preliminary analysis of pain in the rst 15 minutes was done by use of t tests and analysis of covariance. The nal analysis of pain was compared by use of a longitudinal random-coefcients model to account for the correlation of pain assessments within a patient and to model the differential trends across time. The within-patient correlation was accounted for by a rst-order autocorrelation. The primary outcome was a comparison of pain at 24 hours and at 48 hours. Analyses were performed with SAS software (version 9.1; SAS Institute, Cary, NC). The 48-hour pain diaries showed patterns of missing data that were consistent with the observation that pain assessments may not have been recorded while the patient slept. Therefore 2 analyses were performed, one treating the missing data as ignorable and the other inputting pain scores of 0 during sleep.

measurements within the 4 observation periods during the rst hour and only 1 missing VAS value during the rst 4 hours. However, during hours 9 through 20, over 33% of the measurements were missing, and there were 2 hours in the middle of this period when over 60% of the patients did not provide pain measurements. This cycle was then repeated when there were nearly complete VAS measurements provided in hours 24 through 33 and over 33% were missing in hours 34 through 44. All of the patients provided a VAS assessment of pain at 48 hours. This pattern of missing data is common with pain diaries and is the result of the pain measurements not being recorded while patients are asleep. However, it is important to acknowledge any missing values and take appropriate steps to deal with this in the data analysis. Missing values are usually considered to be missing at random. However, this was not the case here. There was clearly a cyclical pattern of missing data consistent with a sleep-wake cycle, with the prilocaine-lidocaine group having a higher percentage of missing data than the eugenol group (P .036). To deal with this issue, 2 analyses were done. The rst analysis compared the trends of the 2 treatment groups across time using the observed data, and the second analysis was done inputting a pain score of 0 whenever the observed data were missing, assuming that the patient did not have pain while asleep. The model used to compare the groups was a longitudinal random-coefcients model to account for the correlation of pain assessments within a patient and to model the differential trends across time. A cubic model of the trends was chosen because it was evident that pain did not exhibit a simple linear trend and the third-order model t the shape of the response prole in each group. The response proles

Results
The mean pretreatment pain score was 6.72 on a scale ranging from 1 to 10 for the eugenol group and 6.37 for the prilocaine-lidocaine group (SE, 0.46), and the 2 groups were not different (P .62). In the immediate post-treatment period (0-15 minutes), the pain levels were signicantly reduced in both groups (Ps .001). However, the thermosetting gel produced a reduction in pain to a mean score of 3.23 (SE, 0.62) whereas the eugenol resulted in signicantly less pain reduction, to a mean score of 4.83 (SE, 0.43) (P .022). In the subsequent 48 hours the pain diary VAS measurements were recorded at 76% of the time points in the eugenol group (787 of 1,040) and in 68% of the time points in the prilocaine-lidocaine mixture group (533 of 780). There were no missing pain

FIGURE 2. VAS pain scores over time in the 2 groups with input data. Burgoyne et al. Topical Anesthetic and Alveolar Osteitis. J Oral Maxillofac Surg 2010.

BURGOYNE ET AL

147

Table 1. ESTIMATED VAS PAIN LEVELS ACROSS TIME IN THE 2 TREATMENT GROUPS

Eugenol Time Estimates without input observations for missing values 6h 12 h 18 h 24 h 30 h 36 h 42 h 48 h Estimates with pain input as 0 for missing observations 6h 12 h 18 h 24 h 30 h 36 h 42 h 48 h
Estimate 95% CI

Prilocaine-Lidocaine Mixture
Estimate 95% CI

Difference Between Eugenol and Prilocaine-Lidocaine Mixture

Estimate P Value 95% CI

3.82 3.15 3.03 3.23 3.53 3.69 3.49 2.69 3.34 2.35 2.02 2.10 2.36 2.57 2.49 1.87

2.83 to 4.80 1.99 to 4.32 1.88 to 4.19 2.19 to 4.28 2.59 to 4.47 2.79 to 4.59 2.58 to 4.39 1.61 to 3.77 2.54 to 4.15 1.51 to 3.20 1.18 to 2.85 1.32 to 2.88 1.64 to 3.09 1.88 to 3.27 1.78 to 3.19 1.78 to 2.79

3.24 2.91 2.75 2.71 2.72 2.72 2.66 2.49 2.58 1.19 1.67 1.68 1.78 1.81 1.59 0.95

2.10 to 4.38 1.56 to 4.26 1.42 to 4.09 1.50 to 3.91 1.63 to 3.81 1.68 to 3.76 1.61 to 3.71 1.22 to 3.76 1.65 to 3.51 0.94 to 2.89 0.71 to 2.64 0.79 to 2.58 0.95 to 2.62 1.01 to 2.60 0.77 to 2.40 0.10 to 2.01

0.58 0.24 0.28 0.53 0.81 0.97 0.82 0.21 0.76 0.44 0.34 0.41 0.58 0.76 0.90 0.92

.439 .786 .751 .507 .259 .163 .235 .803 .214 .497 .588 .484 .296 .150 .097 .190

0.93 to 2.09 1.54 to 2.02 1.49 to 2.04 1.07 to 2.12 0.63 to 2.25 0.41 to 2.35 0.56 to 2.21 1.46 to 1.87 0.46 to 1.99 0.86 to 1.73 0.93 to 1.62 0.78 to 1.60 0.53 to 1.68 0.29 to 1.82 0.17 to 1.97 0.48 to 2.31

NOTE. By use of a longitudinal random-coefcients model, the 2 groups were not signicantly different at 24 hours or at 48 hours (Ps .19). Abbreviation: CI, condence interval.
Burgoyne et al. Topical Anesthetic and Alveolar Osteitis. J Oral Maxillofac Surg 2010.

for each group were plotted and are shown in Figure 1 for the observed data and Figure 2 for the data with missing values input as 0. It should be noted that the pain in the prilocaine-lidocaine group was always nominally less than that in the eugenol group. In addition, the prilocaine-lidocaine mixture and eugenol groups were compared every 6 hours to observe how the response proles changed over time. The results are shown in Table 1. The predicted VAS pain level and 95% condence intervals are shown for each group and time point, as well as the differences and P values for the differences, which are uncorrected for multiple comparisons. It can be observed that there was nominally less pain in the prilocaine-lidocaine group at each time point but that this difference was not statistically signicant. The primary comparison of pain at 24 hours and at 48 hours was also not signicantly different (Ps .19). However, the prilocaine-lidocaine group required a mean of 7.9 analgesic tablets per patient over a period of 48 hours, whereas the eugenol group required a mean of 10.3 tablets per patient. Moreover, over the course of the 48 hours, the medication requirement of the prilocaine-lidocaine group decreased whereas that in the eugenol group increased.

Discussion
The intraoral use of a combination of lidocaine and prilocaine in a thermosetting gel for providing topical anesthesia before scaling and root planning has been described by a number of authors.12,13 However, this is the rst study to use this mixture for the treatment of AO. When such a medication is applied topically, there is always the question of absorption rate and toxicity. Vickers et al14 showed that when 8 g of a 5% lidocaineprilocaine mixture was applied to an 18-cm2 area of buccal mucosa for 30 minutes, the maximum concentrations measured for lidocaine (418 ng/mL) and prilocaine (223 ng/mL) were well below toxic levels. Moreover, Friskopp and Huledal15 showed that when 3.5 g of a lidocaine-prilocaine mixture in a thermosetting gel was placed in the gingival sulcus, the peak plasma concentrations of lidocaine (99-266 ng/mL) and prilocaine (46188 ng/mL) were also not toxic. An interesting property of a lidocaine-prilocaine mixture, shown in a recent study by Berg et al,16 is that it is antibacterial. They exposed 25 strains of bacteria, including Staphylococcus aureus, Methicillin-resistant Staphylococcus aureus, Escherichia coli, Pseudomonas aeruginosa, and Streptococcus pyogenes, to either 2.5% lidocaine2.5% prilocaine or 1% lidocaine and

148 found that the lidocaine-prilocaine mixture had a rapidacting and powerful antibacterial effect whereas the lidocaine alone had only a slight effect on bacterial growth. Thus the use of the lidocaine-prilocaine mixture in AO may have a double effect, reducing both pain and bacterial contamination. Although this study did not show that the thermosetting anesthetic gel was signicantly better than the eugenol in controlling the pain from AO, it was just as effective and its use may have several advantages. First, it is easier for the clinician to use and more comfortable for the patient than placement of an agent with a gauze carrier. Second, it does not require removal. Third, its antibacterial effect may improve the rate of healing. Although the difference between the 2 groups in this study was not statistically signicant at 24 or 48 hours, this may have been because of the relatively small number of patients. Therefore a study with a larger number of patients seems warranted. Acknowledgment
Assistance in the analysis of the data was provided by Al M. Best, PhD, Department of Biostatistics, Virginia Commonwealth University.

TOPICAL ANESTHETIC AND ALVEOLAR OSTEITIS

3. Giglio JA, Rowland RW, Laskin DM, et al: The use of sterile versus non-sterile gloves during out-patient exodontia. Quintessence Int 24:543, 1993 4. Halpern L, Dodson T: Does prophylactic administration of systemic antibiotics prevent postoperative inammatory complications after third molar surgery? J Oral Maxillofac Surg 65:177, 2007 5. Ren Y, Malmstrom H: Effectiveness of antibiotic prophylaxis in third molar surgery: A meta-analysis of randomized controlled clinical trials. J Oral Maxillofac Surg 65:1909, 2007 6. Lynch DP, Newland HR, McClendon J: Myospherulosis of the oral hard and soft tissues. J Oral Maxillofac Surg 42:349, 1984 7. Moore JW, Brekke JH: Foreign body giant cell reaction related to placement of tetracycline-treated polylactic acid: Report of 18 cases. J Oral Maxillofac Surg 48:808, 1990 8. Rutledge JL, Marcoot TM: Terra-Cotril/Gelfoam for reduction of the incidence of localized osteitis following mandibular third molar removal. J Oral Med 39:51, 1984 9. Zuniga JR, Leist JC: Topical tetracycline-induced neuritis: A case report. J Oral Maxillofac Surg 53:196, 1995 10. Betts NJ, Makowski G, Shen YH, et al: Evaluation of topical viscous 2% lidocaine jelly as an adjunct during the management of alveolar osteitis. J Oral Maxillofac Surg 53:1140, 1995 11. Donaldson D, Meechan JG: A comparison of the effects of EMLA cream and topical 5% lidocaine on discomfort during gingival probing. Anesth Prog 42:7, 1995 12. Jeffcoat MK, Geurs NC, Harris PA, et al: Intrapocket anesthesia for scaling and root planning: Results of a double-blind multicenter trial using lidocaine prilocaine dental gel. J Periodontol 72:895, 2001 13. Magnusson I, Guers NC, Harris PA, et al: Intrapocket anesthesia for scaling and root planning in pain-sensitive patients. J Periodontol 74:597, 2003 14. Vickers ER, Marzbani N, Gerzina TM, et al: Pharmacokinetics of EMLA cream 5% application to oral mucosa. Anesth Prog 44:32, 1997 15. Friskopp J, Huledal G: Plasma levels of lidocaine and prilocaine after application of Oraqix, a new intrapocket anesthetic, in patients with advanced periodontitis. J Clin Periodontol 28: 425, 2001 16. Berg JO, Mossner BK, Skov MN, et al: Antibacterial properties of EMLA and lidocaine in wound tissue biopsies for culturing. Wound Repair Regen 1:581, 2006

References
1. Friedman JW: The prophylactic extraction of third molars: A public health hazard. Am J Public Health 97:554, 2007 2. Blum IR: Contemporary views on dry socket (alveolar osteitis): A clinical appraisal of standardization, aetiopathogenesis and management: A critical review. Int J Oral Maxillofac Surg 31: 309, 2002